Hereditary nonpolyposis colon cancer (HNPCC) accounts for 15-20 percent of colon cancers or approximately 23,000 new cases of cancer per year in the United States alone. HNPCC represents the highest incidence of a known hereditary cancer in the human population. Dr. Fishel has recently shown that HNPCC is associated with alteration of the human post- replication mismatch repair genes hMSH2, hMLH1 and hPMS1 which cosegregate with the disease and are its likely cause with a genetic penetrance of better than 90 percent. Furthermore, it appears that up to 30 percent of sporadic tumors of all types manifest the genetic instability that is a hallmark of a human mismatch repair defect. This genetic instability (termed: mutator) provides a fundamentally new way to view the development of cancer. hMSH2, hMLH1 and hPMS1 are related to highly conserved families of mismatch repair proteins identified from bacteria to man. The function of these genes in bacteria is well documented and provided a foundation for the study of the human homologues. The Specific Aims of this proposal are: I. Overproduce, purify and characterize hMSH2, hMLH1 and hPMS1 using well defined assay systems; II. Determine the function and interaction of hMSH2, hMLH1, and hPMS1 by controlled and random mutagenesis and peptide inhibition studies in vitro and in vivo; III. Determine the alterations in function conferred by known HNPCC and spontaneous mutations that lead to human tumors; and IV. Begin the process of reconstituting the human mismatch reaction in vitro using purified components. These studies are straight-forward and should provide the framework for bioassays that can be used to assess the functional alterations associated with human tumors that involve mismatch repair defects. Accumulation of a database on the functional alterations of hMSH2, hMLH1 and hPMS1 should create a firm foundation for diagnostic and therapeutic efficacy.